地震救援机器人的地理环境动态觉察功能优化设计

Optimization Design of Dynamic Detection Function of Geographic Environment for Earthquake Rescue Robot

为了让地震救援机器人能够准确觉察动态地理环境,正常执行救援,研究地震救援机器人的地理环境动态觉察功能优化方法。该方法采用三维激光雷达感知地理环境信息并生成三维地图;通过最小二乘支持向量机算法,分类生成的三维地图,实时分类救援地形;利用改进遗传蚁群算法,规划地震后救援路径。测试结果表明:当信息素因子和启发信息因子分别为0.7和0.9时,该方法能够有效估计机器人位姿,准确避开障碍物,获取多个救援路径规划结果,保证救援的正常进行。

In order to accurately perceive the dynamic geographical environment of the earthquake rescue robot and carry out the rescue normally,the geodynamic environment perception function optimization method of the earthquake rescue robot is studied.This method uses 3D lidar to sense the geographical environment information and generate 3D maps,optimize the improved least squares support vector set algorithm,classify the 3D maps in real time,and plan the post-earthquake rescue path.The test results show that when the pheromone factor and the heuristic information factors are 0.7 and 0.9,respectively,this method can effectively estimate the position,accurately avoid obstacles,and obtain the results of multiple rescue path planning,so as to ensure the normal progress of the rescue.